One of my goals for this year is to spend more time and effort developing effective Visualizations for my various research projects, in an effort to make my research more accessible to others. This is one thing that I think many academics are particularly bad at: letting others know what they are up to, and why it might be something worth looking at. In order to avoid this pitfall, I plan to focus on producing interactive, web-based visuals suitable for a more general audience in addition to more traditional forms of research dissemination such as journals and conference papers. It is my hope that by doing this, I will be making my research more readily available to those who might actually be able to use it, and maybe even create some compelling Visualizations in the process. While I’m not quite ready to start creating full-blown interactive websites yet, I thought it might be a good idea to start with something small to get the ball rolling; so I put together an upgraded version of my previous map of visitors to www.carsonfarmer.com.
map, this time providing some basic interaction with the data. Most of
the functionality is based on examples from the D3 website, and at
this point, the map is really more of a learning tool than anything. As
in my previous static map, the IP addresses were geocoded using the
Data Science Toolkit API via the RDSTK R package, and all data
processing and manipulation was done using R. Additionally, the colour
scheme functionality comes from slide 25 of this presentation,
and uses the sequential colour palettes from colorbrewer.org. There
are still a few kinks to work out (like how to get the
to work properly in Internet Explorer), and tonnes of additional
features and functions could be added, so comments and suggestions are
welcome. Having said that, I think the new version looks quite nice, and
will likely form the basis for more complex visuals as I become more
familiar with Javascipt and various other web-development tools.